CN101520878A - Method, device and system for pushing advertisements to users - Google Patents
Method, device and system for pushing advertisements to users Download PDFInfo
- Publication number
- CN101520878A CN101520878A CN200910081151A CN200910081151A CN101520878A CN 101520878 A CN101520878 A CN 101520878A CN 200910081151 A CN200910081151 A CN 200910081151A CN 200910081151 A CN200910081151 A CN 200910081151A CN 101520878 A CN101520878 A CN 101520878A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- user
- advertisements
- users
- interest
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000005295 random walk Methods 0.000 claims abstract description 29
- 239000013598 vector Substances 0.000 claims description 35
- 239000011159 matrix material Substances 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 28
- 230000005012 migration Effects 0.000 claims description 22
- 238000013508 migration Methods 0.000 claims description 22
- 238000004422 calculation algorithm Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 17
- 230000004044 response Effects 0.000 claims description 9
- 238000003064 k means clustering Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 abstract description 17
- 230000006399 behavior Effects 0.000 description 13
- 230000007704 transition Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- NVNLLIYOARQCIX-MSHCCFNRSA-N Nisin Chemical compound N1C(=O)[C@@H](CC(C)C)NC(=O)C(=C)NC(=O)[C@@H]([C@H](C)CC)NC(=O)[C@@H](NC(=O)C(=C/C)/NC(=O)[C@H](N)[C@H](C)CC)CSC[C@@H]1C(=O)N[C@@H]1C(=O)N2CCC[C@@H]2C(=O)NCC(=O)N[C@@H](C(=O)N[C@H](CCCCN)C(=O)N[C@@H]2C(NCC(=O)N[C@H](C)C(=O)N[C@H](CC(C)C)C(=O)N[C@H](CCSC)C(=O)NCC(=O)N[C@H](CS[C@@H]2C)C(=O)N[C@H](CC(N)=O)C(=O)N[C@H](CCSC)C(=O)N[C@H](CCCCN)C(=O)N[C@@H]2C(N[C@H](C)C(=O)N[C@@H]3C(=O)N[C@@H](C(N[C@H](CC=4NC=NC=4)C(=O)N[C@H](CS[C@@H]3C)C(=O)N[C@H](CO)C(=O)N[C@H]([C@H](C)CC)C(=O)N[C@H](CC=3NC=NC=3)C(=O)N[C@H](C(C)C)C(=O)NC(=C)C(=O)N[C@H](CCCCN)C(O)=O)=O)CS[C@@H]2C)=O)=O)CS[C@@H]1C NVNLLIYOARQCIX-MSHCCFNRSA-N 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000004309 nisin Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method, a device and a system for recommending advertisements to users. The method mainly comprises the steps: clicking information of the advertisements by users is obtained, the users are divided into different groups according to the clicking information of the advertisements by the users, the users in the same group have the general characteristic for the advertisements, the interest degree of the users in the same group for the advertisements is calculated by adopting a random walk model by aiming at each group, and the matched advertisements are selected at least according to the interest degree of the users for the advertisements and are recommended to the users. The invention can sufficiently use the clicking data of the prior advertisements by the users in the process that the random walk model is led into the interest degree of the users, dig the commonness among the users and obtain the potential interests of the users for the advertisements, individuation advertisements can be recommended to the users in a driving mode under the condition that the users do not input information, and the advertising precision of the advertisements is increased.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method, an apparatus, and a system for pushing an advertisement to a user.
Background
The personalized advertisement is a very important means for accurate advertisement delivery, because the recommended advertisement is closely related to the interest of the user, the user experience can be effectively improved, the user can accept the personalized advertisement, from the perspective of an advertiser, the personalized advertisement is purposeful, the waste of advertisement delivery is avoided, and the efficiency of advertisement delivery is improved.
The traditional personalized advertisement recommendation technology is based on user browsing records and behaviors, establishes an interest model for a user, and then matches advertisements with the interest model. The general method for establishing the user interest model is to count webpage information browsed by a user within a period of time, then extract keywords in the webpages browsed by the user or classify the webpages browsed by the user, and take the extracted keywords or the webpage classification result as the characteristics of the user; correspondingly, characteristics are extracted for the advertisements, namely, each advertisement is classified when the advertisements are input, corresponding keywords are set for the advertisements, and the category to which the advertisements belong or the keywords corresponding to the advertisements can be used as the characteristics of the advertisements. When the advertisement needs to be delivered to the user, the characteristics of the advertisement are matched with the characteristics of the user, the advertisement which is most matched with the characteristics of the user is selected, and the matched advertisement is delivered to the user in various modes.
In the prior art, a scheme for advertisement delivery to a user includes: the method comprises the steps of extracting information characteristics of a user from input information of the user, extracting advertisement characteristics containing the information characteristics from an advertisement library, determining advertisement information corresponding to the advertisement characteristics, extracting output information corresponding to the input information of the user from the information library, and pushing the output information and the advertisement information to the corresponding user.
In the process of implementing the present invention, the inventor finds out that in the existing advertisement push scheme for users: only under the condition that the user inputs information, the personalized advertisement can be actively pushed to the user, in other words, the application scene is limited, and a scheme which is not limited by the application scene and can push the personalized advertisement to the user is urgently needed in the industry at present, namely, the scheme can push the personalized advertisement to the user no matter whether the user inputs the information or not.
Disclosure of Invention
The embodiment of the invention provides a method for pushing advertisements to a user, a user interest computing device and an advertisement pushing system, which are used for pushing personalized advertisements to the user under the condition of not being limited by an application scene.
A method for pushing advertisements to users comprises the following steps:
acquiring click information of a user on an advertisement;
dividing the users into different groups according to the clicking information of the users on the advertisements, so that the users in the same group have the common characteristic on the advertisements;
aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model;
and selecting matched advertisements according to the interest degree of the user for the advertisements, and pushing the advertisements to the user.
A user interest computing device, comprising:
the user grouping module is used for grouping the users into different groups according to the obtained click information of the users on the advertisements, so that the users in the same group have the common characteristic on the advertisements;
and the interest degree acquisition module is used for calculating the interest degree of the users in the group on the advertisement by adopting a random walk model aiming at each group.
An advertisement push system comprising:
the user interest calculation device is used for dividing the users into different groups according to the obtained click information of the users on the advertisements, so that the users in the same group have common characteristics on the advertisements; aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model;
and the advertisement pushing device is used for selecting the matched advertisement at least according to the interest degree of the user for the advertisement and pushing the advertisement to the user.
According to the technical scheme provided by the embodiment of the invention, the users are divided into different groups according to the click information of the users on the advertisements, the random walk model is introduced into the calculation process of the interest degrees of the users in the groups on the advertisements, the click data of the users on the existing advertisements is fully utilized to mine the commonality of the users and the users, the interest degrees of the users on the clicked advertisements are obtained, the potential interest degrees of the users on the un-clicked advertisements are also obtained, and the matched advertisements are selected and pushed to the users at least according to the interest degrees of the users on the advertisements, so that the personalized advertisements are actively pushed to the users under the condition that the users do not input information, and the personalized advertisements are pushed to the users under the condition that the application scenes are not limited.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Fig. 1 is a schematic flowchart of a method for pushing an advertisement to a user according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a method for pushing an advertisement to a user according to a second embodiment of the present invention;
FIG. 3 is a schematic diagram of a user and an advertisement according to a second embodiment of the present invention;
fig. 4 is a flowchart illustrating a method for pushing an advertisement to a user according to a third embodiment of the present invention;
FIG. 5 is a schematic structural diagram of a user interest calculation apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an advertisement delivery system according to an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of an advertisement push system according to an embodiment of the present invention when there is no user input information;
fig. 8 is a schematic structural diagram of an advertisement push system according to an embodiment of the present invention when a user inputs information.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
The processing flow of the method for pushing the advertisement to the user provided by the embodiment of the invention is shown in fig. 1, and comprises the following steps:
and 11, acquiring click information of the user on the advertisement.
And 12, dividing the users into different groups according to the click information of the users to the advertisements, so that the users in the same group have common characteristics to the advertisements.
Specifically, the users may be divided into different groups by using a clustering algorithm according to the click information of the users on the advertisements.
Here, the grouping of users into different groups using a clustering algorithm includes: according to the click information of each user on the advertisement, constructing the interest vector of each user, and according to the category of the advertisement, constructing the feature vectors of various advertisements. And clustering the users by adopting a K-means clustering algorithm and dividing the users into different groups by taking the interest vector set cooperation of each user as a set to be clustered, taking the number of the advertisement types as the initial number of the clustering categories, taking the feature vectors of various advertisements as the initial center. It should be understood that it is a necessary condition in the K-means clustering algorithm that the interest vector sets of the respective users cooperate as a set to be clustered, the feature vectors of the various types of advertisements serve as initial centers, and the number of the types of advertisements serve as initial numbers of the clustering categories. By clustering users by adopting the K-means clustering algorithm, users with common characteristics to advertisements can be clustered into the same group, namely, users with relatively high common characteristics are clustered together.
And step 13, aiming at each group, calculating the interest degree of the users in the group on the advertisement by adopting a random walk model.
The method specifically comprises the following steps: calculating the migration probability among all nodes in a first set, wherein the nodes in the first set comprise all users in the group and advertisements clicked by the users;
forming a first matrix by using the migration probability among the nodes, wherein the first matrix is N rows and N columns, N is the total number of the nodes in the first set, elements of i rows and k columns of the first matrix are the calculated migration probability from the ith node to the kth node in the set, and the value ranges of i and k are from 1 to N;
performing iterative processing on the first matrix until the matrix converges, wherein an element on an ith row and a kth column in the converged matrix is an interest degree of an ith node in the kth node, and the iterative processing comprises the following steps:
when the ith node is a user and the kth node is an advertisement, the elements on the ith row and kth column represent the interest degree of the user i in the advertisement kth;
when the ith node is an advertisement and the kth node is a user, the elements on the ith row and kth column represent the attraction of the advertisement i to the user k;
when the ith and kth nodes are all users, the elements on the ith row and kth column represent the similarity between the users i and k;
when the ith and kth nodes are all advertisements, the elements on the ith row and kth column represent the similarity between the advertisements i and k.
And step 14, selecting the matched advertisement at least according to the interest degree of the user for the advertisement, and pushing the advertisement to the user.
Further, the interest level of different users in different advertisements may be stored.
1. When no user inputs information, the specific process of pushing the advertisement to the user in step 14 includes:
when a triggering condition for pushing advertisements to the user is met, ordering the advertisements according to the interest degrees of the user, and pushing a certain number of first series of advertisements which are matched with the interest degrees in the top order to the corresponding user; or,
when a triggering condition for pushing the advertisements to the users is met, selecting a first series of advertisements of which the interest degrees of the users to the advertisements meet a preset interest threshold value, and pushing the first series of advertisements to the corresponding users; or,
when the triggering condition for pushing the advertisements to the users is met, aiming at the appointed advertisement types, sorting the advertisements according to the interest degrees of the users, and pushing a certain number of first series of advertisements which are matched with the interest degrees in the top sorting to the corresponding users; or,
when the triggering condition for pushing the advertisements to the users is met, a first series of advertisements with the interest degree of the advertisements meeting a preset interest threshold value are selected according to the appointed advertisement types, and the advertisements are pushed to the corresponding users.
It should be understood that, if the user inputs the request information, the foregoing step of pushing to the corresponding user further includes:
extracting user information characteristics from request information input by a user, and calculating the content relevance between the user information characteristics and advertisement characteristics of each advertisement in the first series of advertisements;
and determining the advertisements matched with the user according to the content relevance and the interestingness corresponding to the first series of advertisements.
2. When there is information input by the user, the specific process of pushing the advertisement to the user in step 14 includes:
extracting user information characteristics from request information input by a user, and calculating the content relevance between the user information characteristics and advertisement characteristics of each advertisement in an advertisement library;
selecting a certain number of advertisements with content relevance ranked in the front or advertisements with content relevance meeting a preset relevant threshold as content relevant advertisements corresponding to the request information;
and according to the stored interest degree information of the user for the advertisement, acquiring the interest degree of the user for the advertisement related to the content, determining the advertisement matched with the user according to the content relevance degree and the interest degree corresponding to the advertisement related to the content, and returning the matched advertisement and the response information of the user request information to the user.
Under one implementation, determining an advertisement matched with a user according to the content relevance and the interestingness corresponding to the content-relevant advertisement specifically includes:
and calculating the matching degree P of each advertisement in the content-related advertisements to the user according to the content relevance degree and the interestingness corresponding to the content-related advertisements, wherein the advertisements with the certain number are ranked in the front and the matching degree P is a certain number of advertisements, or the advertisements with the matching degree P meeting a preset matching threshold value are the advertisements matched with the user.
Therefore, in the embodiment of the invention, the users are divided into different groups according to the click information of the users on the advertisements, the random walk model is introduced into the calculation process of the interest degrees of the users in the groups on the advertisements, the click data of the users on the existing advertisements is fully utilized to mine the commonality between the users, the interest degrees of the users on the clicked advertisements are obtained, the potential interest degrees of the users on the non-clicked advertisements are also obtained, and the matched advertisements are selected and pushed to the users at least according to the interest degrees of the users on the advertisements, so that the personalized advertisements can be actively pushed to the users no matter whether the users input the information or not, namely the personalized advertisements are pushed to the users without being limited by application scenes.
Example two
The processing flow of the method for pushing the advertisement to the user provided by the embodiment of the invention is shown in fig. 2, and comprises the following processing steps:
and step 21, collecting user access data to obtain the click information of the user on the advertisement.
In practical application, the telecom operator can act as an advertisement service provider, and the placement and the release of the advertisement are managed by the telecom operator. The access data of the user is extracted from telecom operation equipment, such as GGSN (Gateway GPRS supporting Node)/DSLAM (Digital Subscriber line access Multiplexer)/WAP GW (wireless application Protocol Gateway), where the access data may be access information of a user accessing the internet via a mobile phone to a network, including a mobile phone number of the user, an access URL, access time, and the like.
Analyzing the access data to obtain click information of the user on the advertisement, wherein the click information comprises: user information, click time, click on an advertisement, etc. And storing the click information of the user on the advertisement in a user behavior database.
The structure of a user behavior database is shown in table 1 below:
table 1:
| user information | Click on advertisement | Time of click |
| 1394512＊＊＊＊ | http://wap.chinatelecom.com.cn/ad？id＝D01223 | 2004-12-1212:01:12 |
| ... | ... | ... |
And step 22, dividing the users into different groups by adopting a clustering algorithm according to the click information of the users on the advertisements.
After the click information of the users to the advertisements is obtained, the users are divided into different groups by adopting a clustering algorithm, so that the users in the same group have the common characteristic to the advertisements. The specific grouping process is as follows:
according to the records of the users clicking the advertisement, an interest vector of each user is respectively constructed, for example, the interest vector of the ith user can be expressed as: { ni1，...nij，...，nisIn which n isijIndicating the number of clicks of the ith user on the jth advertisement.
Respectively constructing initial gathering points of various advertisements according to the categories of the advertisements, wherein the initial gathering points are represented as eachFeature vectors of class ads. For example, a feature vector for a certain type of advertisement may be represented as {0, 0.,. n., n }i，…，nj0, 0}, where n isi＝...＝njThe i to j advertisements belong to this category, i.e., the advertisement belonging to this category is 1.
The method comprises the steps of taking interest vector sets of all users as a set to be clustered, taking initial aggregation points of various advertisements as initial centers, taking the number of advertisement types as initial numbers of clustering categories, clustering the users by using a k-means clustering algorithm, and obtaining a series of user groups.
The K-means clustering algorithm firstly calculates the distance between each user vector and each initial center respectively, and how many groups there are initial centers, the distance adopts a cosine included angle calculation method, then the users are divided into the group with the central point closest to the distance, and after all the users are grouped, the first iteration is completed; and then recalculating the central point of each group, regrouping the users by using the new central point until the class center is not changed any more, and ending the iteration.
The formula in which the center point of each group is recalculated is as follows:
ujis the new center point of the j sets,|Cji is the number of j sets of vectors, x is the number of CjThe vector of (1).
The users are divided into different groups through the clustering process, so that the calculation amount of the interest degree of the users to the advertisements by adopting a random walk model is reduced, and the clustering is to cluster the users with high commonality to the advertisements, and the random walk model evaluates the potential interest of the users to the advertisements by utilizing the commonality among the users, so that the random walk model can be more effective through the clustering process.
And step 23, calculating interest degrees of the users in each group to the advertisements by using the random walk model in the group unit.
Taking a group of user data, and calculating the interest degree of the group of users for the advertisement by adopting a random walk model, wherein the specific calculation process comprises the following steps:
first, a bipartite graph of the group of users and advertisements is created, as shown in fig. 3, the bipartite graph of the group of users and advertisements is shown, where the left dot represents a user, the right dot represents an advertisement, the solid line represents a user click on the advertisement, the number of clicks may be represented by the number of the continuous lines, and the dotted line corresponds to a user and an advertisement that are not clicked.
A probability of migration between the user and the advertisement is calculated. The bipartite graph shown in fig. 3 is expanded, and the user and the advertisement are both regarded as nodes in the set S, and the transition probability between the nodes in the set S is calculated.
The conventional calculation formula of the transition probability P (i | k) from the point i to the point k is as follows:
N in the above formula 1ikThe number of connecting lines from the point i to the point k is represented, alpha represents the migration probability of the pointed node, the migration probability is a constant, the value range of j is 1 to N, and N is the total number of the nodes in the set S.
The transition probabilities P (i | k) have different meanings according to the meaning represented by i and k:
when one of i and k is an advertisement and the other is a user, the migration probability P (i | k) represents the interest degree of the user in the advertisement or the attraction of the advertisement to the user;
when i and k are both users, the migration probability P (i | k) represents the similarity between the users;
when i, k are both advertisements, the transition probability P (i | k) represents the similarity between advertisements.
The calculated transition probability when i and k are equal has no specific meaning, but for the convenience of later calculation, the value is alpha, 0< alpha <1, and corresponding processing is carried out on the condition that i and k are not equal, and the value is multiplied by a coefficient 1-alpha.
The interest level P (a |3) of the user a in the advertisement 3 is calculated according to the calculation formula of P (i | k), and we see from fig. 3 that the user a does not directly click on the advertisement 3, so the obtained transition probability P (a |3) is 0, which indicates that a is not interested in 3. But the actual situation may not be, we can see that the user a clicks on the advertisement 2, and b clicks on both 2 and 3, which indicates that there is a commonality between the a and b points and, if the commonality is large enough, or the commonality is sufficient, for example, the users a and b are both interested in amazon shopping web and computer books, and both click on the advertisement related to amazon shopping web, and if the user b clicks on the advertisement related to the current shopping web, it is known that the user a is also interested in the current shopping web, but the interest level is lower than that of the amazon shopping web, and in order to calculate the potential interest of the user in the advertisement that has not been clicked, the embodiment uses a random walk model to calculate the interest level of the user in the advertisement, where the interest level of the user in the advertisement includes: the user's interest level in the clicked advertisement and the user's potential interest level in the advertisement that has not been clicked.
From fig. 3, we can see that although point a cannot reach point 3 directly, point a can reach point 2 first, then reach point b, and finally reach point 3 through point b, that is, point a >2- > b- >3, and point a can reach point 3 after walking 3 steps. It can be seen that the more interesting user a is for point 3, the more lines can reach point 3 by this constant wandering.
Mathematically the relationship between the points changes after any n-step walk can be calculated by matrix iteration. In this embodiment, the transition probability calculated according to the above formula 1 constitutes a matrix a, where i rows and k columns of elements in a are the above transition probability P (i | k), and the matrix a is N rows and N columns. Iterating the matrix A for n times (n is the eigenvector of the matrix A) until convergence is reached, wherein the element P' (i | k) on the ith row and k column is the interest level of the node i on the node k, that is, the interest level of the node i on the node k is obtained
When the node i is a user and the node k is an advertisement, the P' (i | k) represents the interest degree of the user i in the advertisement k;
when the node i is an advertisement and the node k is a user, the P' (i | k) represents the attraction of the advertisement i to the user k;
when the nodes i and k are both users, the P' (i | k) represents the similarity between the users i and k;
when both nodes i, k are advertisements, the P' (i | k) represents the similarity between the advertisements i, k.
Similarly, the element P' (i | j) in the ith row and j column is the interest level of the user i in the advertisement j.
After the interest degrees of the user i for all the advertisements are obtained, the interest degree information of the user i for each advertisement is stored in a user interest database according to the ascending order or the descending order of the interest degrees. The structure of a user interest database is shown in table 2 below:
table 2:
| user identification | Advertisement sign | Degree of interest |
| 1394512**** | D01223 | 0.254 |
| ... | ... | ... |
It should be noted that, if the telecommunication operator acts as an advertisement service provider, and the advertisement delivery and publication are managed by the telecommunication operator, the telecommunication operator stores all the information of the advertisement, including the link of the advertisement, the id of the advertisement, and so on, so that the id of the advertisement can be obtained through the link of the advertisement. In another mode, the telecom operator may also cooperate with a partner (a content provider or a network operator), the advertisement publishing management is charged by the partner, the telecom operator provides data services, collects feedback information of the user and submits the feedback information to the partner for processing, and the partner obtains a history access record of the user and a click record of the advertisement and recommends the advertisement to the user by using the processing flow of the embodiment of the present invention.
The above processing is repeatedly executed until all the user groups are processed, and the interest degrees of all the users in all the groups in the advertisement can be obtained and stored in the user interest database.
The process of calculating the interest degree of each group of users in the advertisement by using the random walk model may be triggered at regular time or when the collected user behavior data reaches a certain threshold value.
And 24, selecting the matched advertisement at least according to the interest degree of the user for the advertisement, and pushing the advertisement to the user.
When an operator determines that an advertisement needs to be pushed to a user according to a certain trigger condition, the trigger condition may be that a request sent by the user is received, or a predetermined timing time for the user is reached, or a predetermined condition for the user is met, for example, the user and a telecommunication operator sign an advertisement allowing to send catering discount information, and when the telecommunication operator obtains a large amount of catering information advertisements, the advertisement is pushed to the user.
Extracting the interest degree information of the user on various advertisements stored in the user interest database, sorting the advertisements according to the interest degrees of the user on the advertisements, and pushing a certain number of advertisements matched with the interest degrees sorted in the front to the corresponding users; or,
selecting advertisements with interest degrees of users for the advertisements meeting a preset interest threshold value, and pushing the advertisements to corresponding users; or,
aiming at the appointed advertisement type, ordering according to the interest degree of the user for the advertisement type, and pushing a certain number of advertisements which are matched with the interest degree in the top order to the corresponding user; or,
and selecting advertisements with interest degrees of users for the advertisements meeting a preset interest threshold value aiming at the appointed advertisement types, and pushing the advertisements to the corresponding users.
The advertisement can be pushed by sending an email with the advertisement to the user, or sending an advertisement short message to the user, and so on.
Therefore, in the embodiment, the clustering algorithm and the random walk model are introduced into the process of the interest degree of the user in the advertisement, the common characteristics of the user and the user are mined by fully utilizing the click data of the user on the existing advertisement, the interest degree of the user on the clicked advertisement is obtained, the potential interest degree of the user on the non-clicked advertisement is also obtained, and the matched advertisement is selected and pushed to the user at least according to the interest degree of the user on the advertisement, so that the advertisement is actively pushed to the user under the condition that the user does not input information, and the advertisement putting precision is improved.
EXAMPLE III
According to the embodiment of the invention, the interest degree of the user for the advertisement is obtained according to the processing flow provided by the second embodiment, and the interest degree is stored in the user interest database. And searching related advertisements according to the information input by the user, and calculating the content relevance between the searched advertisements and the information input by the user. And the matched advertisements are pushed to the user by comprehensively considering the interest degree and the content relevance degree of the user on the advertisements. The processing flow of the method for pushing the advertisement to the user provided by the embodiment is shown in fig. 4, and includes the following processing steps:
Specifically, the method for receiving and acquiring information input by a user, based on keyword or information classification, extracts user information features from the request information, calculates the similarity between the user information features and advertisement features of each advertisement in an advertisement library, the similarity is also called content relevance, and takes a certain number of advertisements with the similarity arranged in the front or advertisements with the content relevance meeting a preset relevant threshold as content relevant advertisements corresponding to the request information, namely advertisements relevant to the user information content.
One method for calculating the content relevancy is as follows: and constructing a user information characteristic vector I ═ I1.,. in } and an advertisement characteristic vector A ═ a 1.,. an }, calculating a cosine included angle of the two vectors, and obtaining a similarity p1 between the user information characteristic and the advertisement characteristic.
For example, a user accessing the Internet with a mobile phone requests an ISP (Internet Service Provider) to access a news webpage of a certain sports category, and after receiving the request of the user, the ISP analyzes the user and the request of the user, and retrieves a pile of advertisements related to sports from an advertisement library.
And step 42, obtaining the interest degree of the user for the content-related advertisement from a user interest database.
Specifically, the user interest library is queried according to the advertisement identifier and the user identifier of the content-related advertisement obtained in step 41, so as to obtain the interest level of the user in the content-related advertisement.
And 43, calculating the matching degree of the advertisement according to the content relevance degree and the interestingness information corresponding to the content-related advertisement, and determining the advertisement which is most matched with the user according to the matching degree.
Calculating the matching degree P of the user to the content-related advertisement, wherein the calculating method of the matching degree P comprises the following steps:
where P1 is the content relevance of the content-relevant advertisement, P2 is the user's interest level in the content-relevant advertisement,
for adjusting the factor, can be adjusted byThe proportion of the content relevance and the interestingness in the calculation of the matching degree is adjusted,take between (1, 0).
And selecting a certain number of advertisements with the top-ranked matching degree as the advertisements which are most matched with the user.
And step 44, sending the advertisement which is matched with the user most and response information of the user request information to the user.
When response information needs to be returned to the user according to the information input by the user, the advertisement needing to be pushed to the user is inserted into the response information and is returned to the user. Such as into a web page requested by the user or into a video requested by the user.
When the user receives the response message, the user also sees the advertisement pushed to the user.
It should be noted that, in the embodiment of the present invention, the order of selecting the content-related advertisement and selecting the advertisement that the user is interested in may be changed, for example, after obtaining the first batch of matched advertisements by referring to step 34 in the second embodiment, the content-related degrees corresponding to the first batch of advertisements are further obtained, the matching degree of the advertisements is calculated according to the interest degree and the content-related degree corresponding to the first batch of advertisements, and the advertisement that is most matched with the user is determined from the first batch of advertisements according to the matching degree.
Therefore, in the embodiment, under the condition that the user inputs information, the content relevance of the retrieved advertisements and the interest degree of the user in the advertisements can be comprehensively considered, so that the best matched advertisements are screened out and pushed to the user.
An embodiment of the present invention further provides a user interest calculating device, a structure of which is shown in fig. 5, and the structure may include:
the user grouping module 51 is used for grouping the users into different groups according to the obtained click information of the users on the advertisements, so that the users in the same group have common characteristics on the advertisements;
and the interestingness acquiring module 52 is used for calculating the interestingness of the users in the group to the advertisement by adopting a random walk model aiming at each group.
Under one implementation, the user grouping module 51 specifically includes:
the vector construction module 511 is configured to construct interest vectors of each user according to click information of the user on the advertisement, which is obtained from the collected user access data, and to construct feature vectors of various types of advertisements according to categories to which the advertisements belong;
and a clustering processing module 512, configured to cluster the users by using the interest vector set of each user obtained by the vector construction module as a cluster set, using the feature vectors of each type of advertisement obtained by the vector construction module as an initial center, using the number of the advertisement types as an initial number of a clustering category, and using a K-means clustering algorithm to cluster the users, so as to divide the users into different groups.
In one implementation, the interestingness obtaining module 52 specifically includes:
and the inter-node migration probability calculation module 521 is configured to calculate a migration probability between nodes in the first set, where the nodes in the first set include all users in the group and advertisements clicked by the users.
The method is specifically used for: selecting a group of users, using each user in the group of users as a node, using each advertisement as a node, forming all the nodes into a first set, and calculating the migration probability among the nodes in the first set.
An interest degree calculating module 522, configured to use the migration probabilities among the nodes calculated by the inter-node migration probability calculating module to form a first matrix, and perform iterative processing on the first matrix until the first matrix converges, where elements on an ith row and a k column in the converged matrix are interest degrees of an ith node to a kth node, where the first matrix is N rows and N columns, where N is the total number of nodes in the set, elements on an i row and a k column in the first matrix are the calculated migration probabilities of the ith node to the kth node in the set, and values of i and k range from 1 to N.
Specifically, when the ith node is a user and the kth node is an advertisement, the elements in the ith row and kth column represent the interest degree of the user i in the advertisement k;
when the ith node is an advertisement and the node k is a user, the elements on the ith row and k columns represent the attraction of the advertisement i to the user k;
when the ith and kth nodes are all users, the elements on the ith row and kth column represent the similarity between the users i and k;
when the ith and kth nodes are all advertisements, the elements on the ith row and kth column represent the similarity between the advertisements i and k.
Therefore, in the user interest calculation device in the embodiment of the invention, the users are divided into different groups according to the click information of the users on the advertisements, and the random walk model is introduced into the calculation process of the interest degree of the users on the advertisements in each group, so that the click data of the users on the existing advertisements is fully utilized to mine the commonality between the users, and not only the interest degree of the users on the clicked advertisements is obtained, but also the potential interest degree of the users on the non-clicked advertisements is obtained.
An embodiment of the present invention further provides an advertisement delivery system, the structure of which is shown in fig. 6, including:
the user interest calculation device 61 is used for dividing the users into different groups according to the obtained click information of the users on the advertisements, so that the users in the same group have common characteristics on the advertisements; aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model;
and the advertisement pushing device 62 is used for selecting the matched advertisement according to at least the interest degree of the user in the advertisement and pushing the advertisement to the user.
In one implementation, the advertisement delivery system of the embodiment of the present invention further includes a user interest database, configured to store the interest degrees of different users for different advertisements calculated by the user interest calculation device;
the advertisement pushing device 62 is a first advertisement pushing device, and is configured to, when a trigger condition for pushing an advertisement to a user is met, rank the advertisements according to interest degrees of the user, and push a certain number of advertisements, which are matched with the interest degrees ranked in the top, to the corresponding user;
or, when the triggering condition for pushing the advertisement to the user is met, selecting the advertisement of which the interest degree of the user to the advertisement meets a preset interest threshold value, and pushing the advertisement to the corresponding user;
or when the triggering condition for pushing the advertisements to the users is met, ordering the advertisements according to the interest degrees of the users in the type of the appointed advertisements according to the interest degrees of the users, and pushing a certain number of advertisements which are matched with the interest degrees in the top order to the corresponding users;
or, when the trigger condition for pushing the advertisement to the user is met, selecting the advertisement with the interest degree of the user for the advertisement meeting the preset interest threshold value according to the appointed advertisement type, and pushing the advertisement to the corresponding user.
In another implementation, the advertisement delivery system of the embodiment of the present invention further includes a user interest database, configured to store the interest degrees of different users for different advertisements, which are calculated by the user interest calculation device;
the content-related advertisement determining device is used for extracting user information characteristics from request information input by a user and calculating the content relevance between the user information characteristics and advertisement characteristics of each advertisement in an advertisement library, wherein the content relevance ranks a certain number of advertisements in the front, or the advertisements with the content relevance meeting a preset relevant threshold value are the content-related advertisements related to the request information;
the advertisement pushing device 62 is a second advertisement pushing device, and is configured to obtain the interest level of the user in the content-related advertisement, determine an advertisement matched with the user according to the content-related advertisement and the interest level corresponding to the content-related advertisement, and return the matched advertisement and response information of the user request information to the user.
The following describes a specific implementation of the advertisement delivery system according to an embodiment of the present invention with reference to specific drawings:
1. when no user inputs information, a specific implementation structure of the advertisement delivery system is shown in fig. 7, and includes: information collection means 70, user interest calculation means 71, advertisement push means 72, and user behavior database 73, user interest database 74, user database 75, and advertisement database 76. Wherein,
a user database 75 for storing user detailed data including a user identifier and the like;
an advertisement database 76 for storing advertisement detail data containing advertisement identification and the like;
and an information collecting device 70 for collecting the access data of the user, obtaining the click information of the user on the advertisement according to the access data, and storing the click information of the user on the advertisement in the user behavior database 73.
A user behavior database 73 is structured as shown in table 1 above.
The user interest calculation device 71 is used for acquiring the click information of the user on the advertisement from the user behavior database 73, and dividing the users into different groups by adopting a clustering algorithm according to the click information so that the users in the same group have common characteristics on the advertisement; aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model; and stores the calculated user interest level in the advertisement in the user interest database 74.
The structure of a user interest database is shown in table 2 above.
The advertisement pushing device 72 is configured to, when a trigger condition for pushing the advertisement to the user is satisfied, sort the advertisements in the user interest database according to the interest degrees of the users in the advertisements, obtain a certain number of advertisements, which are matched with the interest degrees sorted in the front, from the advertisement database 76, and push the advertisements to the corresponding users;
or, when the trigger condition for pushing the advertisement to the user is satisfied, the advertisement with the interest degree of the user for the advertisement satisfying the preset interest threshold is obtained from the advertisement database 76, and the advertisement is pushed to the corresponding user;
or, when the trigger condition for pushing the advertisement to the user is satisfied, for the appointed advertisement type, sorting the advertisement type in the user interest database 74 according to the interest degree of the user for the advertisement type, acquiring a certain number of advertisements matched with the interest degree in the front sorting from the advertisement database 76, and pushing the advertisements to the corresponding user;
or, when the trigger condition for pushing the advertisement to the user is met, the advertisement with the interest degree of the advertisement meeting the preset interest threshold value is extracted from the advertisement database 76 according to the appointed advertisement type, and the advertisement is pushed to the corresponding user. It should be understood that different advertisement push functions can be flexibly adopted according to actual needs.
2. When there is information input by a user, a specific implementation structure of the advertisement delivery system is shown in fig. 8, and includes: information collecting means 80, user interest calculating means 81, content-related advertisement determining means 82 and advertisement pushing means 83, and user behavior database 84, user interest database 85, user database 86, advertisement database 87 and information database 88; the user behavior database 84, the user interest database 85, and the user database 86 are the same as above, and therefore are not described again;
an advertisement database 87 for storing detailed advertisement data including advertisement identifiers and advertisement characteristics, the advertisement characteristics being generally identified by categories to which advertisements belong and keywords;
an information database 88 for storing information requested by the user.
And the information collecting device 80 is used for collecting the access data of the user, obtaining the click information of the user on the advertisement according to the access data, and storing the click information of the user on the advertisement into the user behavior database 84.
The user interest calculation device 81 is used for acquiring the click information of the user on the advertisement from the user behavior database 84, and dividing the users into different groups by adopting a clustering algorithm according to the click information so that the users in the same group have common characteristics on the advertisement; aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model; and stores the calculated user's interest level in the advertisement in the user interest database 85.
A content-related advertisement determination device 82, configured to extract user information features from request information input by a user, extract advertisement features of each advertisement from an advertisement database 87, and calculate a content relevance between the user information features and each advertisement feature, where a certain number of advertisements with top-ranked content relevance in the advertisement database 87, or advertisements with content relevance satisfying a preset relevance threshold are content-related advertisements related to the request information;
and the advertisement pushing device 83 is configured to obtain the interest degree of the user for the content-related advertisement from the user interest database 85, determine an advertisement matched with the user according to the content-related degree and the interest degree corresponding to the content-related advertisement, and return the matched advertisement and response information of the user request information to the user.
Therefore, in the advertisement push system of the embodiment of the invention, users are divided into different groups according to the click information of the users on the advertisements, the random walk model is introduced into the calculation process of the interest degrees of the users in the groups on the advertisements, the click data of the users on the existing advertisements is fully utilized to mine the commonality of the users and the users, the interest degrees of the users on the clicked advertisements and the potential interest degrees of the users on the un-clicked advertisements are obtained through the commonality of the users and the users, and the matched advertisements are selected and pushed to the users at least according to the interest degrees of the users on the advertisements, so that the personalized advertisements can be actively pushed to the users no matter whether the users input information or not, namely, the personalized advertisements are pushed to the users under the condition of not being limited by application scenes, and the potential interest of the users on the advertisements is calculated, thereby improving the accuracy of advertisement delivery.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
In summary, the embodiments of the present invention introduce the clustering algorithm and the random walk model into the process of the interest degree of the user in the advertisement, so that the click data of the user on the existing advertisement can be fully utilized, and the click data of the user on the advertisement is used as the main reference factor of the interest of the user. The method and the device can fully mine the commonality between the user and the user, can acquire the potential interest of the user to the advertisement, actively push the advertisement to the user under the condition that the user does not input information, and improve the advertisement putting precision.
The embodiment of the invention can comprehensively consider the content relevance of the searched content-related advertisement and the interest of the user in the content-related advertisement under the condition that the user inputs information, obtain the advertisement which is most matched with the user, and push the advertisement to the user.
According to the embodiment of the invention, the users are grouped through a clustering algorithm, and then the interest degree of the users in the group to the advertisement is calculated, so that the random walk model is optimized, and the problems of large calculation amount and long time consumption of the random walk model are solved.
The embodiment of the invention not only can be used for guiding the advertisement putting, but also can be used for measuring the advertisement publishing effect, fully excavates the potential clicking behavior of the user by calculating the interest degree of the user in the advertisement, and is more objective and fair than simply counting the clicking rate of the user on the advertisement to evaluate the advertisement putting effect.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (15)
1. A method for advertisement push to a user, comprising:
acquiring click information of a user on an advertisement;
dividing the users into different groups according to the clicking information of the users on the advertisements, so that the users in the same group have the common characteristic on the advertisements;
aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model;
and selecting matched advertisements according to the interest degree of the user for the advertisements, and pushing the advertisements to the user.
2. The method of claim 1, wherein the grouping users into different groups according to the click information of the users on the advertisement comprises:
and according to the click information of the users to the advertisements, dividing the users into different groups by adopting a clustering algorithm.
3. The method of claim 2, wherein the grouping users into different groups according to the click information of the users on the advertisement by using a clustering algorithm comprises:
constructing interest vectors of all users according to the click information of all users on the advertisements;
constructing feature vectors of various advertisements according to the categories to which the advertisements belong;
the method comprises the steps of taking the cooperation of interest vector sets of all users as a set to be clustered, taking feature vectors of various advertisements as initial centers, taking the number of advertisement types as the initial number of clustering categories, clustering the users by adopting a K-means clustering algorithm, and dividing the users into different groups.
4. The method for pushing advertisement to user according to claim 1, wherein said calculating interest level of users in the group for advertisement by using random walk model comprises:
calculating the migration probability among all nodes in a first set, wherein the nodes in the first set comprise all users in the group and advertisements clicked by the users;
forming a first matrix by using the migration probability among the nodes, wherein the first matrix is N rows and N columns, N is the total number of the nodes in the first set, elements of i rows and k columns of the first matrix are the calculated migration probability from the ith node to the kth node in the set, and the value ranges of i and k are from 1 to N;
and performing iterative processing on the first matrix until the matrix is converged, wherein elements on an ith row and an ith column in the converged matrix are the interest degree of an ith node in the kth node.
5. The method of advertisement push to user according to claim 4, characterized by:
when the ith node is a user and the kth node is an advertisement, the elements on the ith row and kth column represent the interest degree of the user i in the advertisement kth;
when the ith node is an advertisement and the kth node is a user, the elements on the ith row and kth column represent the attraction of the advertisement i to the user k;
when the ith and kth nodes are all users, the elements on the ith row and kth column represent the similarity between the users i and k;
when the ith and kth nodes are all advertisements, the elements on the ith row and kth column represent the similarity between the advertisements i and k.
6. The method of any of claims 1 to 5, wherein the method further comprises: storing the interest degrees of different users for different advertisements;
the selecting and pushing the matched advertisement according to the interest degree of the user to the advertisement comprises:
when a triggering condition for pushing advertisements to the user is met, ordering the advertisements according to the interest degrees of the user, and pushing a certain number of first series of advertisements which are matched with the interest degrees in the top order to the corresponding user; or,
selecting a first series of advertisements of which the interest degrees of the users to the advertisements meet a preset interest threshold value, and pushing the first series of advertisements to the corresponding users; or,
aiming at the appointed advertisement type, ordering according to the interest degree of the user for the advertisement type, and pushing a certain number of first series of advertisements which are matched with the interest degree in the top order to the corresponding user; or,
and aiming at the appointed advertisement type, selecting a first series of advertisements of which the interest degrees of the users to the advertisements meet a preset interest threshold value, and pushing the advertisements to the corresponding users.
7. The method of any claim 6, wherein if the user inputs the request information, the step of pushing further comprises:
extracting user information characteristics from request information input by a user, and calculating the content relevance between the user information characteristics and advertisement characteristics of each advertisement in the first series of advertisements;
and determining the advertisements matched with the user according to the content relevance and the interestingness corresponding to the first series of advertisements.
8. The method of any of claims 1 to 5, wherein the method further comprises: storing the information of interest degrees of different users for different advertisements; the step of selecting the matched advertisement according to at least the interest degree of the user in the advertisement and pushing the advertisement to the user comprises the following steps:
extracting user information characteristics from request information input by a user, and calculating the content relevance between the user information characteristics and advertisement characteristics of each advertisement in an advertisement library;
selecting a certain number of advertisements with content relevance ranked in the front or advertisements with content relevance meeting a preset relevant threshold as content relevant advertisements corresponding to the request information;
and according to the stored interest degree information of the user for the advertisement, acquiring the interest degree of the user for the advertisement related to the content, determining the advertisement matched with the user according to the content relevance degree and the interest degree corresponding to the advertisement related to the content, and returning the matched advertisement and the response information of the user request information to the user.
9. The method of claim 8, wherein determining the advertisement matching the user according to the content relevance and the interest level corresponding to the content-relevant advertisement comprises:
and calculating the matching degree P of each advertisement in the content-related advertisements to the user according to the content relevance degree and the interestingness corresponding to the content-related advertisements, wherein the advertisements with the certain number are ranked in the front and the matching degree P is a certain number of advertisements, or the advertisements with the matching degree P meeting a preset matching threshold value are the advertisements matched with the user.
10. A user interest computing apparatus, comprising:
the user grouping module is used for grouping the users into different groups according to the obtained click information of the users on the advertisements, so that the users in the same group have the common characteristic on the advertisements;
and the interest degree acquisition module is used for calculating the interest degree of the users in the group on the advertisement by adopting a random walk model aiming at each group.
11. The apparatus of claim 10, wherein the user grouping module comprises:
the vector construction module is used for constructing interest vectors of all users according to click information of the users on the advertisements, which is obtained from the collected user access data, and constructing characteristic vectors of various advertisements according to the categories to which the advertisements belong;
and the clustering processing module is used for clustering the users by adopting a mean clustering algorithm by taking the interest vector set of each user obtained by the vector construction module as a cluster set, taking the number of the advertisement types as the initial number of the clustering categories and taking the feature vectors of various advertisements obtained by the vector construction module as the initial center, and dividing the users into different groups.
12. The apparatus for pushing advertisement to user according to claim 10, wherein the interestingness obtaining module comprises:
an inter-node migration probability calculation module, configured to calculate a migration probability between nodes in a first set, where the nodes in the first set include all users in the group and advertisements clicked by the users;
an interest degree calculation module, configured to use the migration probabilities among the nodes calculated by the inter-node migration probability calculation module to form a first matrix, perform iterative processing on the first matrix until the first matrix converges, where elements on an ith row and a kth column in the converged matrix are interest degrees of an ith node to a kth node, where the first matrix is N rows and N columns, where N is the total number of nodes in the set, elements on an ith row and a kth column of the first matrix are the calculated migration probabilities of the ith node to the kth node in the set, and values of i and k range from 1 to N.
13. An advertisement push system, comprising:
the user interest calculation device is used for dividing the users into different groups according to the obtained click information of the users on the advertisements, so that the users in the same group have common characteristics on the advertisements; aiming at each group, calculating the interest degree of the users in the group to the advertisement by adopting a random walk model;
and the advertisement pushing device is used for selecting the matched advertisement at least according to the interest degree of the user for the advertisement and pushing the advertisement to the user.
14. The system of claim 13, further comprising a user interest database for storing the interest level of different users in different advertisements calculated by the user interest calculation device;
the advertisement pushing device is a first advertisement pushing device and is used for sorting the advertisements according to the interest degrees of the users when the triggering condition for pushing the advertisements to the users is met, and pushing a certain number of advertisements which are matched with the interest degrees in the front sorting to the corresponding users;
or, when the triggering condition for pushing the advertisement to the user is met, selecting the advertisement of which the interest degree of the user to the advertisement meets a preset interest threshold value, and pushing the advertisement to the corresponding user;
or when the triggering condition for pushing the advertisements to the users is met, ordering the advertisements according to the interest degrees of the users in the type of the appointed advertisements according to the interest degrees of the users, and pushing a certain number of advertisements which are matched with the interest degrees in the top order to the corresponding users;
or, when the trigger condition for pushing the advertisement to the user is met, selecting the advertisement with the interest degree of the user for the advertisement meeting the preset interest threshold value according to the appointed advertisement type, and pushing the advertisement to the corresponding user.
15. The system of claim 13, further comprising a user interest database for storing the interest level of different users in different advertisements calculated by the user interest calculation device;
the advertisement push system further comprises:
the content-related advertisement determining device is used for extracting user information characteristics from request information input by a user and calculating the content relevance between the user information characteristics and advertisement characteristics of each advertisement in an advertisement library, wherein the content relevance ranks a certain number of advertisements in the front, or the advertisements with the content relevance meeting a preset relevant threshold value are the content-related advertisements related to the request information;
the advertisement pushing device is a second advertisement pushing device and is used for obtaining the interest degree of the user for the content-related advertisement, determining the advertisement matched with the user according to the content-related degree and the interest degree corresponding to the content-related advertisement, and returning the matched advertisement and the response information of the user request information to the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910081151A CN101520878A (en) | 2009-04-03 | 2009-04-03 | Method, device and system for pushing advertisements to users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910081151A CN101520878A (en) | 2009-04-03 | 2009-04-03 | Method, device and system for pushing advertisements to users |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101520878A true CN101520878A (en) | 2009-09-02 |
Family
ID=41081454
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910081151A Pending CN101520878A (en) | 2009-04-03 | 2009-04-03 | Method, device and system for pushing advertisements to users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101520878A (en) |
Cited By (93)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011079462A1 (en) * | 2009-12-31 | 2011-07-07 | Nokia Corporation | Methods and apparatuses for user interest modeling |
| CN102238045A (en) * | 2010-04-27 | 2011-11-09 | 广州迈联计算机科技有限公司 | System and method for predicting user behavior in wireless Internet |
| CN102289503A (en) * | 2011-08-24 | 2011-12-21 | 百度在线网络技术（北京）有限公司 | Online application searching method and system |
| CN102377790A (en) * | 2010-08-04 | 2012-03-14 | 腾讯科技（深圳）有限公司 | Method and device for pushing data |
| CN102402766A (en) * | 2011-12-27 | 2012-04-04 | 纽海信息技术（上海）有限公司 | User interest modeling method based on webpage browsing |
| CN102521767A (en) * | 2011-12-13 | 2012-06-27 | 亿赞普(北京)科技有限公司 | Method and system for publishing network advertising information |
| CN102592223A (en) * | 2011-01-18 | 2012-07-18 | 卓望数码技术（深圳）有限公司 | Commodity recommending method and commodity recommending system |
| CN102668309A (en) * | 2009-11-17 | 2012-09-12 | 高通股份有限公司 | Condition-based wireless power |
| CN102708131A (en) * | 2011-03-02 | 2012-10-03 | 奥多比公司 | Automatic classification of consumers into micro-segments |
| CN102739655A (en) * | 2012-06-12 | 2012-10-17 | 成都边界元科技有限公司 | Information continuous pushing method of multiple interest points |
| CN102779139A (en) * | 2011-05-13 | 2012-11-14 | 百度在线网络技术（北京）有限公司 | Method and device for determining representation forms in information promotional technology of search engine |
| CN102956014A (en) * | 2011-08-12 | 2013-03-06 | 源初科技有限公司 | Method of attention-targeting for online advertisement |
| CN102999589A (en) * | 2012-11-15 | 2013-03-27 | Tcl集团股份有限公司 | Method and system for recommending order |
| WO2013078640A1 (en) * | 2011-11-30 | 2013-06-06 | Google Inc. | Estimating user demographics |
| CN103150663A (en) * | 2013-02-18 | 2013-06-12 | 亿赞普(北京)科技有限公司 | Method and device for placing network placement data |
| CN103164508A (en) * | 2011-12-15 | 2013-06-19 | 国际商业机器公司 | Systems and method for customizing browsing content according to user data |
| CN103257963A (en) * | 2012-02-15 | 2013-08-21 | 珠海德百祺科技有限公司 | Method and system for acquiring information |
| CN103279878A (en) * | 2013-05-22 | 2013-09-04 | 广州威信信息科技发展有限公司 | Precise interactive digital television advertisement putting system and method |
| CN103577478A (en) * | 2012-08-06 | 2014-02-12 | 腾讯科技（深圳）有限公司 | Web page pushing method and system |
| CN103593786A (en) * | 2013-08-14 | 2014-02-19 | 北京世纪喜鹊信息科技有限公司 | Advertisement orientation dimension delivering network hardware device based on Internet |
| CN103824214A (en) * | 2014-03-17 | 2014-05-28 | 联想(北京)有限公司 | Information processing method and device and electronic equipment |
| US8853995B2 (en) | 2009-06-12 | 2014-10-07 | Qualcomm Incorporated | Devices for conveying wireless power and methods of operation thereof |
| WO2014161452A1 (en) * | 2013-04-02 | 2014-10-09 | Tencent Technology (Shenzhen) Company Limited | System and method for pushing and distributing promotion content |
| WO2014180407A1 (en) * | 2013-12-06 | 2014-11-13 | 中兴通讯股份有限公司 | Pushing method and device therefor |
| CN104217353A (en) * | 2014-08-20 | 2014-12-17 | 苏州海云网络科技有限公司 | Webpage advertisement directional pushing system |
| CN104361062A (en) * | 2014-11-03 | 2015-02-18 | 百度在线网络技术（北京）有限公司 | Associated information recommendation method and device |
| CN104391883A (en) * | 2014-11-05 | 2015-03-04 | 北京航空航天大学 | Online advertisement audience sorting method based on transfer learning |
| CN104574134A (en) * | 2013-10-23 | 2015-04-29 | 传播公司 | Method, system and computer-readable recording medium for associating advertisement with specific user group |
| CN104615707A (en) * | 2015-02-02 | 2015-05-13 | 百度在线网络技术（北京）有限公司 | Information point determining method and device |
| CN104657506A (en) * | 2015-03-13 | 2015-05-27 | 百度在线网络技术（北京）有限公司 | Data processing method and device based on user scenario |
| CN104754526A (en) * | 2013-12-30 | 2015-07-01 | 中国移动通信集团上海有限公司 | Method for distributing short message and server |
| CN104901817A (en) * | 2014-03-07 | 2015-09-09 | 腾讯科技（北京）有限公司 | Method and device for pushing target information |
| US9177327B2 (en) | 2011-03-02 | 2015-11-03 | Adobe Systems Incorporated | Sequential engine that computes user and offer matching into micro-segments |
| CN105095219A (en) * | 2014-04-23 | 2015-11-25 | 华为技术有限公司 | Weibo recommendation method and terminal |
| CN105550903A (en) * | 2015-12-25 | 2016-05-04 | 腾讯科技（深圳）有限公司 | Target user determination method and apparatus |
| CN105721528A (en) * | 2014-12-04 | 2016-06-29 | 中国移动通信集团江苏有限公司 | Local area network content push method, device and system |
| CN105812844A (en) * | 2014-12-29 | 2016-07-27 | 深圳市Tcl高新技术开发有限公司 | User advertisement push method and system of TV |
| CN105809464A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Method and device for information delivery |
| US9454763B2 (en) | 2010-08-24 | 2016-09-27 | Adobe Systems Incorporated | Distribution of offer to a social group by sharing based on qualifications |
| CN102289503B (en) * | 2011-08-24 | 2016-12-14 | 百度在线网络技术（北京）有限公司 | Application on site searching method and system |
| CN106294410A (en) * | 2015-05-22 | 2017-01-04 | 苏宁云商集团股份有限公司 | A kind of determination method of personalized information push time and determine system |
| CN106355449A (en) * | 2016-08-31 | 2017-01-25 | 腾讯科技（深圳）有限公司 | User selecting method and device |
| CN106412635A (en) * | 2016-09-29 | 2017-02-15 | 北京赢点科技有限公司 | Intelligent advertising method and system |
| WO2017024509A1 (en) * | 2015-08-11 | 2017-02-16 | 常平 | Advertisement push method and advertisement push system |
| WO2017045415A1 (en) * | 2015-09-17 | 2017-03-23 | 腾讯科技（深圳）有限公司 | Content delivery method and device |
| CN106557957A (en) * | 2016-11-30 | 2017-04-05 | 天脉聚源（北京）科技有限公司 | The recommendation method and device of advertisement |
| WO2017071251A1 (en) * | 2015-10-28 | 2017-05-04 | 百度在线网络技术（北京）有限公司 | Information pushing method and device |
| CN106843900A (en) * | 2017-02-09 | 2017-06-13 | 北京奇虎科技有限公司 | The method for pushing and device of a kind of information |
| WO2017101413A1 (en) * | 2015-12-15 | 2017-06-22 | 乐视控股（北京）有限公司 | Information pushing method and information pushing apparatus |
| CN106919946A (en) * | 2015-12-25 | 2017-07-04 | 华为技术有限公司 | A kind of method and device of audience selection |
| WO2017128424A1 (en) * | 2016-01-31 | 2017-08-03 | 胡明祥 | Method for collecting data about advertisement pushing technology, and pushing system |
| CN107038213A (en) * | 2017-02-28 | 2017-08-11 | 华为技术有限公司 | A kind of method and device of video recommendations |
| CN107122805A (en) * | 2017-05-15 | 2017-09-01 | 腾讯科技（深圳）有限公司 | A kind of user clustering method and apparatus |
| CN107318033A (en) * | 2017-06-30 | 2017-11-03 | 深圳市九洲电器有限公司 | Television advertising recommends method and system |
| CN107329988A (en) * | 2017-06-05 | 2017-11-07 | 国政通科技股份有限公司 | A kind of method and system that data are provided |
| CN107341686A (en) * | 2017-05-25 | 2017-11-10 | 段元文 | Divide method, terminal and the computer-readable recording medium of sharp advertisement |
| CN107392655A (en) * | 2017-07-03 | 2017-11-24 | 携程计算机技术（上海）有限公司 | Reward voucher method for pushing, system, storage medium, electronic equipment and shunt method |
| CN107656945A (en) * | 2016-08-25 | 2018-02-02 | 腾讯科技（北京）有限公司 | Information processing method and information processor |
| CN107851260A (en) * | 2015-03-06 | 2018-03-27 | 斯泰瓦纳托集团股份有限公司 | The method and system of information is selectively provided to target group |
| CN107979762A (en) * | 2014-05-08 | 2018-05-01 | 谷歌有限责任公司 | Using live TV stream as ad serving |
| CN107977865A (en) * | 2017-12-07 | 2018-05-01 | 畅捷通信息技术股份有限公司 | Advertisement sending method, device, computer equipment and readable storage medium storing program for executing |
| CN107977382A (en) * | 2016-10-25 | 2018-05-01 | 北京京东尚科信息技术有限公司 | Method and apparatus for pushed information |
| CN107993084A (en) * | 2016-10-27 | 2018-05-04 | 北京酷我科技有限公司 | A kind of advertisement sending method |
| CN108062678A (en) * | 2016-11-08 | 2018-05-22 | 阿里巴巴集团控股有限公司 | A kind of advertisement bit allocation method, device and advertisement delivery system |
| CN108241652A (en) * | 2016-12-23 | 2018-07-03 | 北京国双科技有限公司 | Keyword clustering method and device |
| CN108446927A (en) * | 2011-06-06 | 2018-08-24 | 恩弗伦斯媒体公司 | Consumer drives ad system |
| CN108596694A (en) * | 2012-09-18 | 2018-09-28 | 北京点网聚科技有限公司 | Method, system and respective media for providing a user content |
| CN108628832A (en) * | 2018-05-08 | 2018-10-09 | 中国联合网络通信集团有限公司 | A kind of information keyword acquisition methods and device |
| CN108898432A (en) * | 2018-06-25 | 2018-11-27 | 武汉斗鱼网络科技有限公司 | Advertisement putting effect evaluation method and device and electronic equipment |
| CN108960918A (en) * | 2018-06-29 | 2018-12-07 | 深圳春沐源控股有限公司 | The method and system of information push |
| CN109087711A (en) * | 2018-06-28 | 2018-12-25 | 郑州大学第附属医院 | Medical big data method for digging and system |
| CN109215258A (en) * | 2018-08-10 | 2019-01-15 | 成都贝图科技有限公司 | A kind of books multimode system for the distribution of commodities |
| CN109636445A (en) * | 2018-11-19 | 2019-04-16 | 平安科技（深圳）有限公司 | A kind of advertisement update method, device and terminal device based on user's operation information |
| WO2019128311A1 (en) * | 2017-12-29 | 2019-07-04 | 广东神马搜索科技有限公司 | Advertisement similarity processing method and apparatus, calculation device, and storage medium |
| CN110083777A (en) * | 2018-01-26 | 2019-08-02 | 腾讯科技（深圳）有限公司 | A kind of social network user group technology, device and server |
| CN110263862A (en) * | 2019-06-21 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Method for pushing, device, electronic equipment and the readable storage medium storing program for executing of information |
| CN110288365A (en) * | 2018-03-19 | 2019-09-27 | 北京京东尚科信息技术有限公司 | Data processing method and system, computer system and computer readable storage medium storing program for executing |
| CN110430291A (en) * | 2019-08-02 | 2019-11-08 | 秒针信息技术有限公司 | Data processing method, device, equipment and storage medium |
| CN110730119A (en) * | 2019-10-24 | 2020-01-24 | 珠海格力电器股份有限公司 | Multimedia information pushing method and device and storage medium |
| CN110751526A (en) * | 2019-10-25 | 2020-02-04 | 广州凌鑫达实业有限公司 | Advertisement pushing method and device, computer equipment and storage medium |
| CN110784500A (en) * | 2019-04-02 | 2020-02-11 | 北京嘀嘀无限科技发展有限公司 | Information pushing method and device |
| CN111127095A (en) * | 2019-12-20 | 2020-05-08 | 秒针信息技术有限公司 | Target audience interest analysis method, device, equipment and storage medium |
| CN111127101A (en) * | 2019-12-24 | 2020-05-08 | 北京汇牛科技有限公司 | Advertisement targeted delivery method, device and storage medium |
| CN111385190A (en) * | 2020-03-06 | 2020-07-07 | 杭州威佩网络科技有限公司 | Message sending method and device, electronic equipment and storage medium |
| WO2020143849A3 (en) * | 2020-04-13 | 2021-02-11 | Alipay (Hangzhou) Information Technology Co., Ltd. | Method and system for optimizing user grouping for advertisement |
| CN112581161A (en) * | 2020-12-04 | 2021-03-30 | 上海明略人工智能(集团)有限公司 | Object selection method and device, storage medium and electronic equipment |
| CN112948697A (en) * | 2021-04-01 | 2021-06-11 | 哈尔滨理工大学 | Scientific article recommendation algorithm based on bipartite graph |
| CN112967100A (en) * | 2021-04-02 | 2021-06-15 | 杭州网易云音乐科技有限公司 | Similar population expansion method, device, computing equipment and medium |
| CN113411627A (en) * | 2021-06-17 | 2021-09-17 | 广州博冠信息科技有限公司 | Data pushing method and device, readable storage medium and electronic equipment |
| CN113674012A (en) * | 2020-05-14 | 2021-11-19 | 南宁富桂精密工业有限公司 | Advertisement information pushing method and system |
| CN116417148A (en) * | 2023-05-11 | 2023-07-11 | 纳里健康科技有限公司 | Medical information recommendation method considering user activation interest in 5G application field |
| CN117829916A (en) * | 2023-09-18 | 2024-04-05 | 安徽大学 | App advertisement putting method and system |
| US12039569B2 (en) | 2016-08-25 | 2024-07-16 | Tencent Technology (Shenzhen) Company Limited | Information processing method and device, display terminal, and storage medium |
-
2009
- 2009-04-03 CN CN200910081151A patent/CN101520878A/en active Pending
Cited By (143)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8853995B2 (en) | 2009-06-12 | 2014-10-07 | Qualcomm Incorporated | Devices for conveying wireless power and methods of operation thereof |
| CN102668309A (en) * | 2009-11-17 | 2012-09-12 | 高通股份有限公司 | Condition-based wireless power |
| US9680313B2 (en) | 2009-11-17 | 2017-06-13 | Qualcomm Incorporated | Authorized based receipt of wireless power |
| CN102668309B (en) * | 2009-11-17 | 2016-06-08 | 高通股份有限公司 | Method and apparatus for wireless power transfer |
| US9502909B2 (en) | 2009-11-17 | 2016-11-22 | Qualcomm Incorporated | Power management for electronic devices |
| CN102687166A (en) * | 2009-12-31 | 2012-09-19 | 诺基亚公司 | Methods and apparatuses for user interest modeling |
| CN102687166B (en) * | 2009-12-31 | 2016-02-10 | 诺基亚技术有限公司 | For the method and apparatus of user interest modeling |
| US9031952B2 (en) | 2009-12-31 | 2015-05-12 | Nokia Corporation | Methods and apparatuses for user interest modeling |
| WO2011079462A1 (en) * | 2009-12-31 | 2011-07-07 | Nokia Corporation | Methods and apparatuses for user interest modeling |
| CN102238045A (en) * | 2010-04-27 | 2011-11-09 | 广州迈联计算机科技有限公司 | System and method for predicting user behavior in wireless Internet |
| CN102377790B (en) * | 2010-08-04 | 2016-02-24 | 腾讯科技（深圳）有限公司 | A kind of method and apparatus of propelling data |
| CN102377790A (en) * | 2010-08-04 | 2012-03-14 | 腾讯科技（深圳）有限公司 | Method and device for pushing data |
| US9454763B2 (en) | 2010-08-24 | 2016-09-27 | Adobe Systems Incorporated | Distribution of offer to a social group by sharing based on qualifications |
| CN102592223A (en) * | 2011-01-18 | 2012-07-18 | 卓望数码技术（深圳）有限公司 | Commodity recommending method and commodity recommending system |
| CN102708131A (en) * | 2011-03-02 | 2012-10-03 | 奥多比公司 | Automatic classification of consumers into micro-segments |
| US9177327B2 (en) | 2011-03-02 | 2015-11-03 | Adobe Systems Incorporated | Sequential engine that computes user and offer matching into micro-segments |
| CN102708131B (en) * | 2011-03-02 | 2015-09-09 | 奥多比公司 | By consumer's automatic classification in fine point |
| US10078853B2 (en) | 2011-03-02 | 2018-09-18 | Adobe Systems Incorporated | Offer matching for a user segment |
| CN102779139A (en) * | 2011-05-13 | 2012-11-14 | 百度在线网络技术（北京）有限公司 | Method and device for determining representation forms in information promotional technology of search engine |
| CN108460632A (en) * | 2011-06-06 | 2018-08-28 | 恩弗伦斯媒体公司 | Consumer drives ad system |
| CN108446927A (en) * | 2011-06-06 | 2018-08-24 | 恩弗伦斯媒体公司 | Consumer drives ad system |
| CN102956014A (en) * | 2011-08-12 | 2013-03-06 | 源初科技有限公司 | Method of attention-targeting for online advertisement |
| CN102289503B (en) * | 2011-08-24 | 2016-12-14 | 百度在线网络技术（北京）有限公司 | Application on site searching method and system |
| CN102289503A (en) * | 2011-08-24 | 2011-12-21 | 百度在线网络技术（北京）有限公司 | Online application searching method and system |
| WO2013078640A1 (en) * | 2011-11-30 | 2013-06-06 | Google Inc. | Estimating user demographics |
| CN102521767A (en) * | 2011-12-13 | 2012-06-27 | 亿赞普(北京)科技有限公司 | Method and system for publishing network advertising information |
| CN103164508A (en) * | 2011-12-15 | 2013-06-19 | 国际商业机器公司 | Systems and method for customizing browsing content according to user data |
| CN102402766A (en) * | 2011-12-27 | 2012-04-04 | 纽海信息技术（上海）有限公司 | User interest modeling method based on webpage browsing |
| CN102402766B (en) * | 2011-12-27 | 2017-07-28 | 北京京东尚科信息技术有限公司 | A kind of user interest modeling method based on web page browsing |
| CN103257963A (en) * | 2012-02-15 | 2013-08-21 | 珠海德百祺科技有限公司 | Method and system for acquiring information |
| CN103257963B (en) * | 2012-02-15 | 2017-03-15 | 珠海德百祺科技有限公司 | Information getting method and system |
| CN102739655B (en) * | 2012-06-12 | 2015-04-08 | 成都边界元科技有限公司 | Information continuous pushing method of multiple interest points |
| CN102739655A (en) * | 2012-06-12 | 2012-10-17 | 成都边界元科技有限公司 | Information continuous pushing method of multiple interest points |
| CN103577478A (en) * | 2012-08-06 | 2014-02-12 | 腾讯科技（深圳）有限公司 | Web page pushing method and system |
| CN103577478B (en) * | 2012-08-06 | 2015-07-29 | 腾讯科技（深圳）有限公司 | Web page push method and system |
| CN108596694A (en) * | 2012-09-18 | 2018-09-28 | 北京点网聚科技有限公司 | Method, system and respective media for providing a user content |
| CN102999589B (en) * | 2012-11-15 | 2017-04-19 | Tcl集团股份有限公司 | Method and system for recommending order |
| CN102999589A (en) * | 2012-11-15 | 2013-03-27 | Tcl集团股份有限公司 | Method and system for recommending order |
| CN103150663A (en) * | 2013-02-18 | 2013-06-12 | 亿赞普(北京)科技有限公司 | Method and device for placing network placement data |
| WO2014161452A1 (en) * | 2013-04-02 | 2014-10-09 | Tencent Technology (Shenzhen) Company Limited | System and method for pushing and distributing promotion content |
| US9436768B2 (en) | 2013-04-02 | 2016-09-06 | Tencent Technology (Shenzhen) Company Limited | System and method for pushing and distributing promotion content |
| CN103279878B (en) * | 2013-05-22 | 2016-08-17 | 广东技术师范学院 | A kind of digital television interaction advertisement accurately jettison system and method |
| CN103279878A (en) * | 2013-05-22 | 2013-09-04 | 广州威信信息科技发展有限公司 | Precise interactive digital television advertisement putting system and method |
| CN103593786A (en) * | 2013-08-14 | 2014-02-19 | 北京世纪喜鹊信息科技有限公司 | Advertisement orientation dimension delivering network hardware device based on Internet |
| CN104574134A (en) * | 2013-10-23 | 2015-04-29 | 传播公司 | Method, system and computer-readable recording medium for associating advertisement with specific user group |
| WO2014180407A1 (en) * | 2013-12-06 | 2014-11-13 | 中兴通讯股份有限公司 | Pushing method and device therefor |
| CN104702484A (en) * | 2013-12-06 | 2015-06-10 | 中兴通讯股份有限公司 | Push method and device |
| CN104754526A (en) * | 2013-12-30 | 2015-07-01 | 中国移动通信集团上海有限公司 | Method for distributing short message and server |
| CN104754526B (en) * | 2013-12-30 | 2019-01-18 | 中国移动通信集团上海有限公司 | A kind of method and server for distributing short message |
| CN104901817A (en) * | 2014-03-07 | 2015-09-09 | 腾讯科技（北京）有限公司 | Method and device for pushing target information |
| WO2015131748A1 (en) * | 2014-03-07 | 2015-09-11 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for pushing target information |
| US11196829B2 (en) | 2014-03-07 | 2021-12-07 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for pushing target information |
| JP2017501497A (en) * | 2014-03-07 | 2017-01-12 | テンセント・テクノロジー・（シェンジェン）・カンパニー・リミテッド | Method and apparatus for pushing target information |
| CN103824214A (en) * | 2014-03-17 | 2014-05-28 | 联想(北京)有限公司 | Information processing method and device and electronic equipment |
| CN105095219B (en) * | 2014-04-23 | 2019-02-01 | 华为技术有限公司 | Micro-blog recommendation method and terminal |
| CN105095219A (en) * | 2014-04-23 | 2015-11-25 | 华为技术有限公司 | Weibo recommendation method and terminal |
| CN107979762A (en) * | 2014-05-08 | 2018-05-01 | 谷歌有限责任公司 | Using live TV stream as ad serving |
| CN107979762B (en) * | 2014-05-08 | 2021-01-12 | 谷歌有限责任公司 | Serving live streams as advertisements |
| CN104217353A (en) * | 2014-08-20 | 2014-12-17 | 苏州海云网络科技有限公司 | Webpage advertisement directional pushing system |
| CN104361062A (en) * | 2014-11-03 | 2015-02-18 | 百度在线网络技术（北京）有限公司 | Associated information recommendation method and device |
| CN104361062B (en) * | 2014-11-03 | 2017-10-31 | 百度在线网络技术（北京）有限公司 | A kind of recommendation method and device of related information |
| CN104391883A (en) * | 2014-11-05 | 2015-03-04 | 北京航空航天大学 | Online advertisement audience sorting method based on transfer learning |
| CN104391883B (en) * | 2014-11-05 | 2017-06-20 | 北京航空航天大学 | A kind of online advertisement audient's sort method based on transfer learning |
| CN105721528A (en) * | 2014-12-04 | 2016-06-29 | 中国移动通信集团江苏有限公司 | Local area network content push method, device and system |
| CN105721528B (en) * | 2014-12-04 | 2019-10-22 | 中国移动通信集团江苏有限公司 | A kind of local area network content delivery method, equipment and system |
| CN105812844A (en) * | 2014-12-29 | 2016-07-27 | 深圳市Tcl高新技术开发有限公司 | User advertisement push method and system of TV |
| CN105812844B (en) * | 2014-12-29 | 2019-02-26 | 深圳市Tcl高新技术开发有限公司 | A kind of the user advertising method for pushing and system of TV |
| CN105809464A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Method and device for information delivery |
| CN104615707A (en) * | 2015-02-02 | 2015-05-13 | 百度在线网络技术（北京）有限公司 | Information point determining method and device |
| CN104615707B (en) * | 2015-02-02 | 2018-04-13 | 百度在线网络技术（北京）有限公司 | The definite method and device of information point |
| CN107851260A (en) * | 2015-03-06 | 2018-03-27 | 斯泰瓦纳托集团股份有限公司 | The method and system of information is selectively provided to target group |
| CN104657506A (en) * | 2015-03-13 | 2015-05-27 | 百度在线网络技术（北京）有限公司 | Data processing method and device based on user scenario |
| CN106294410A (en) * | 2015-05-22 | 2017-01-04 | 苏宁云商集团股份有限公司 | A kind of determination method of personalized information push time and determine system |
| WO2017024509A1 (en) * | 2015-08-11 | 2017-02-16 | 常平 | Advertisement push method and advertisement push system |
| WO2017045415A1 (en) * | 2015-09-17 | 2017-03-23 | 腾讯科技（深圳）有限公司 | Content delivery method and device |
| US10621516B2 (en) | 2015-09-17 | 2020-04-14 | Tencent Technology (Shenzhen) Company Limited | Content delivery method, apparatus, and storage medium |
| WO2017071251A1 (en) * | 2015-10-28 | 2017-05-04 | 百度在线网络技术（北京）有限公司 | Information pushing method and device |
| WO2017101413A1 (en) * | 2015-12-15 | 2017-06-22 | 乐视控股（北京）有限公司 | Information pushing method and information pushing apparatus |
| CN105550903A (en) * | 2015-12-25 | 2016-05-04 | 腾讯科技（深圳）有限公司 | Target user determination method and apparatus |
| CN106919946A (en) * | 2015-12-25 | 2017-07-04 | 华为技术有限公司 | A kind of method and device of audience selection |
| WO2017128424A1 (en) * | 2016-01-31 | 2017-08-03 | 胡明祥 | Method for collecting data about advertisement pushing technology, and pushing system |
| CN107656945A (en) * | 2016-08-25 | 2018-02-02 | 腾讯科技（北京）有限公司 | Information processing method and information processor |
| US12039569B2 (en) | 2016-08-25 | 2024-07-16 | Tencent Technology (Shenzhen) Company Limited | Information processing method and device, display terminal, and storage medium |
| WO2018041168A1 (en) * | 2016-08-31 | 2018-03-08 | 腾讯科技（深圳）有限公司 | Information pushing method, storage medium and server |
| CN106355449A (en) * | 2016-08-31 | 2017-01-25 | 腾讯科技（深圳）有限公司 | User selecting method and device |
| CN106355449B (en) * | 2016-08-31 | 2021-09-07 | 腾讯科技（深圳）有限公司 | User selection method and device |
| US11574139B2 (en) | 2016-08-31 | 2023-02-07 | Tencent Technology (Shenzhen) Company Limited | Information pushing method, storage medium and server |
| CN106412635B (en) * | 2016-09-29 | 2019-07-30 | 北京赢点科技有限公司 | A kind of intelligence advertisement placement method and system |
| CN106412635A (en) * | 2016-09-29 | 2017-02-15 | 北京赢点科技有限公司 | Intelligent advertising method and system |
| CN107977382B (en) * | 2016-10-25 | 2022-02-01 | 北京京东尚科信息技术有限公司 | Method and device for pushing information |
| CN107977382A (en) * | 2016-10-25 | 2018-05-01 | 北京京东尚科信息技术有限公司 | Method and apparatus for pushed information |
| CN107993084A (en) * | 2016-10-27 | 2018-05-04 | 北京酷我科技有限公司 | A kind of advertisement sending method |
| CN108062678A (en) * | 2016-11-08 | 2018-05-22 | 阿里巴巴集团控股有限公司 | A kind of advertisement bit allocation method, device and advertisement delivery system |
| CN106557957A (en) * | 2016-11-30 | 2017-04-05 | 天脉聚源（北京）科技有限公司 | The recommendation method and device of advertisement |
| CN108241652A (en) * | 2016-12-23 | 2018-07-03 | 北京国双科技有限公司 | Keyword clustering method and device |
| CN106843900A (en) * | 2017-02-09 | 2017-06-13 | 北京奇虎科技有限公司 | The method for pushing and device of a kind of information |
| CN107038213A (en) * | 2017-02-28 | 2017-08-11 | 华为技术有限公司 | A kind of method and device of video recommendations |
| CN107122805A (en) * | 2017-05-15 | 2017-09-01 | 腾讯科技（深圳）有限公司 | A kind of user clustering method and apparatus |
| CN107341686A (en) * | 2017-05-25 | 2017-11-10 | 段元文 | Divide method, terminal and the computer-readable recording medium of sharp advertisement |
| CN107329988A (en) * | 2017-06-05 | 2017-11-07 | 国政通科技股份有限公司 | A kind of method and system that data are provided |
| CN107318033B (en) * | 2017-06-30 | 2019-11-12 | 深圳市九洲电器有限公司 | Television advertising recommended method and system |
| WO2019001166A1 (en) * | 2017-06-30 | 2019-01-03 | 深圳市九洲电器有限公司 | Television advertisement pushing method, system and electronic device |
| CN107318033A (en) * | 2017-06-30 | 2017-11-03 | 深圳市九洲电器有限公司 | Television advertising recommends method and system |
| CN107392655A (en) * | 2017-07-03 | 2017-11-24 | 携程计算机技术（上海）有限公司 | Reward voucher method for pushing, system, storage medium, electronic equipment and shunt method |
| CN107392655B (en) * | 2017-07-03 | 2020-06-05 | 携程计算机技术（上海）有限公司 | Coupon pushing method, coupon pushing system, storage medium, electronic equipment and distribution method |
| CN107977865A (en) * | 2017-12-07 | 2018-05-01 | 畅捷通信息技术股份有限公司 | Advertisement sending method, device, computer equipment and readable storage medium storing program for executing |
| WO2019128311A1 (en) * | 2017-12-29 | 2019-07-04 | 广东神马搜索科技有限公司 | Advertisement similarity processing method and apparatus, calculation device, and storage medium |
| CN110083777A (en) * | 2018-01-26 | 2019-08-02 | 腾讯科技（深圳）有限公司 | A kind of social network user group technology, device and server |
| CN110083777B (en) * | 2018-01-26 | 2022-11-25 | 腾讯科技（深圳）有限公司 | Social network user grouping method and device and server |
| CN110288365A (en) * | 2018-03-19 | 2019-09-27 | 北京京东尚科信息技术有限公司 | Data processing method and system, computer system and computer readable storage medium storing program for executing |
| CN110288365B (en) * | 2018-03-19 | 2024-09-20 | 北京京东尚科信息技术有限公司 | Data processing method and system, computer system and computer readable storage medium |
| CN108628832A (en) * | 2018-05-08 | 2018-10-09 | 中国联合网络通信集团有限公司 | A kind of information keyword acquisition methods and device |
| CN108898432A (en) * | 2018-06-25 | 2018-11-27 | 武汉斗鱼网络科技有限公司 | Advertisement putting effect evaluation method and device and electronic equipment |
| CN108898432B (en) * | 2018-06-25 | 2022-05-13 | 武汉斗鱼网络科技有限公司 | Advertisement putting effect evaluation method and device and electronic equipment |
| CN109087711A (en) * | 2018-06-28 | 2018-12-25 | 郑州大学第附属医院 | Medical big data method for digging and system |
| WO2020000872A1 (en) * | 2018-06-29 | 2020-01-02 | 深圳春沐源控股有限公司 | Information pushing method and system |
| CN108960918A (en) * | 2018-06-29 | 2018-12-07 | 深圳春沐源控股有限公司 | The method and system of information push |
| CN109215258A (en) * | 2018-08-10 | 2019-01-15 | 成都贝图科技有限公司 | A kind of books multimode system for the distribution of commodities |
| CN109636445A (en) * | 2018-11-19 | 2019-04-16 | 平安科技（深圳）有限公司 | A kind of advertisement update method, device and terminal device based on user's operation information |
| CN110784500A (en) * | 2019-04-02 | 2020-02-11 | 北京嘀嘀无限科技发展有限公司 | Information pushing method and device |
| CN110784500B (en) * | 2019-04-02 | 2021-01-22 | 北京嘀嘀无限科技发展有限公司 | Information pushing method and device, electronic equipment and storage medium |
| CN110263862A (en) * | 2019-06-21 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Method for pushing, device, electronic equipment and the readable storage medium storing program for executing of information |
| CN110263862B (en) * | 2019-06-21 | 2021-05-07 | 北京字节跳动网络技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
| CN110430291A (en) * | 2019-08-02 | 2019-11-08 | 秒针信息技术有限公司 | Data processing method, device, equipment and storage medium |
| CN110730119B (en) * | 2019-10-24 | 2020-12-25 | 珠海格力电器股份有限公司 | Multimedia information pushing method and device and storage medium |
| CN110730119A (en) * | 2019-10-24 | 2020-01-24 | 珠海格力电器股份有限公司 | Multimedia information pushing method and device and storage medium |
| CN110751526A (en) * | 2019-10-25 | 2020-02-04 | 广州凌鑫达实业有限公司 | Advertisement pushing method and device, computer equipment and storage medium |
| CN111127095B (en) * | 2019-12-20 | 2023-05-30 | 秒针信息技术有限公司 | Target audience interest analysis method, device, equipment and storage medium |
| CN111127095A (en) * | 2019-12-20 | 2020-05-08 | 秒针信息技术有限公司 | Target audience interest analysis method, device, equipment and storage medium |
| CN111127101A (en) * | 2019-12-24 | 2020-05-08 | 北京汇牛科技有限公司 | Advertisement targeted delivery method, device and storage medium |
| CN111385190A (en) * | 2020-03-06 | 2020-07-07 | 杭州威佩网络科技有限公司 | Message sending method and device, electronic equipment and storage medium |
| WO2020143849A3 (en) * | 2020-04-13 | 2021-02-11 | Alipay (Hangzhou) Information Technology Co., Ltd. | Method and system for optimizing user grouping for advertisement |
| US11227309B2 (en) | 2020-04-13 | 2022-01-18 | Alipay (Hangzhou) Information Technology Co., Ltd. | Method and system for optimizing user grouping for advertisement |
| CN113674012A (en) * | 2020-05-14 | 2021-11-19 | 南宁富桂精密工业有限公司 | Advertisement information pushing method and system |
| CN112581161A (en) * | 2020-12-04 | 2021-03-30 | 上海明略人工智能(集团)有限公司 | Object selection method and device, storage medium and electronic equipment |
| CN112581161B (en) * | 2020-12-04 | 2024-01-19 | 上海明略人工智能(集团)有限公司 | Object selection method and device, storage medium and electronic equipment |
| CN112948697A (en) * | 2021-04-01 | 2021-06-11 | 哈尔滨理工大学 | Scientific article recommendation algorithm based on bipartite graph |
| CN112967100A (en) * | 2021-04-02 | 2021-06-15 | 杭州网易云音乐科技有限公司 | Similar population expansion method, device, computing equipment and medium |
| CN112967100B (en) * | 2021-04-02 | 2024-03-15 | 杭州网易云音乐科技有限公司 | Similar crowd expansion method, device, computing equipment and medium |
| CN113411627A (en) * | 2021-06-17 | 2021-09-17 | 广州博冠信息科技有限公司 | Data pushing method and device, readable storage medium and electronic equipment |
| CN116417148A (en) * | 2023-05-11 | 2023-07-11 | 纳里健康科技有限公司 | Medical information recommendation method considering user activation interest in 5G application field |
| CN116417148B (en) * | 2023-05-11 | 2024-10-22 | 纳里健康科技有限公司 | Medical information recommendation method considering user activation interest in 5G application field |
| CN117829916A (en) * | 2023-09-18 | 2024-04-05 | 安徽大学 | App advertisement putting method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101520878A (en) | Method, device and system for pushing advertisements to users | |
| CN107526807B (en) | Information recommendation method and device | |
| CN105701216B (en) | A kind of information-pushing method and device | |
| CN103914478B (en) | Webpage training method and system, webpage Forecasting Methodology and system | |
| Lee et al. | MONERS: A news recommender for the mobile web | |
| WO2018041168A1 (en) | Information pushing method, storage medium and server | |
| CN104063801B (en) | A kind of moving advertising recommend method based on cluster | |
| CN106339383B (en) | A kind of search ordering method and system | |
| CN108363821A (en) | A kind of information-pushing method, device, terminal device and storage medium | |
| CN109684538A (en) | A kind of recommended method and recommender system based on individual subscriber feature | |
| US20090276729A1 (en) | Adaptive user feedback window | |
| US20090327913A1 (en) | Using web revisitation patterns to support web interaction | |
| CN102004782A (en) | Search result sequencing method and search result sequencer | |
| US20150205580A1 (en) | Method and System for Sorting Online Videos of a Search | |
| CN101814083A (en) | Automatic webpage classification method and system | |
| CN108876537A (en) | A kind of mixed recommendation method for on-line mall system | |
| CN103731738A (en) | Video recommendation method and device based on user group behavioral analysis | |
| CN104268142B (en) | Based on the Meta Search Engine result ordering method for being rejected by strategy | |
| CN102799591A (en) | Method and device for providing recommended word | |
| CN101401062A (en) | Method and system for determining relevant sources, querying and merging results from multiple content sources | |
| CN107193883B (en) | Data processing method and system | |
| CN107896153B (en) | Traffic package recommendation method and device based on mobile user internet surfing behavior | |
| CN104778237A (en) | Individual recommending method and system based on key users | |
| CN112632405A (en) | Recommendation method, device, equipment and storage medium | |
| CN103198098A (en) | Network information transfer method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20090902 |